Search

I review laboratory research on the development of mass spectrometric methodology for the determination of the structure of natural products of biological and medical interest, which I conducted from 1958 to the end of the twentieth century. The methodology was developed by converting small peptides to their corresponding polyamino alcohols to make them amenable to mass spectrometry, thereby making it applicable to whole proteins. The structures of alkaloids were determined by analyzing the fragmentation of a known alkaloid and then using the results to deduce the structures of related compounds. Heparin-like structures were investigated by determining their molecular weights from the mass of protonated molecular ions of complexes with highly basic, synthetic peptides. Mass spectrometry was also employed in the analysis of lunar material returned by the Apollo missions. A miniaturized gas chromatograph mass spectrometer was sent to Mars on board of the two Viking 1976 spacecrafts.

The development of mass spectrometric methodologies for determining the amino acid sequences of peptides and the primary structure of proteins began in the late 1950s with the conversion of small peptides to analogous polyamino alcohols that were sufficiently volatile to be separated by gas chromatography and ionized by electron impact. Sequencing was greatly simplified by the ability to ionize intact peptides and even entire proteins directly afforded by the invention of FAB, MALDI, and ESI in the 1980s. The advent of DNA sequencing provided further synergistic avenues to the efficient determination of the primary structure of even very large proteins.

A recent paper by Navarro-Gonzalez et al. [Navarro-Gonzalez R, Navarro KF, de la Rosa J, Iniguez E, Molina P, Miranda LD, Morales P, Cienfuegos E, Coll P, Raulin F, et al. (2006) Proc Natl Acad Sci USA 103:16089–16094] claims to show that the Viking GCMS (gas chromatograph–mass spectrometer) experiment, which carried out a search for organic matter at the surface of Mars in 1976, “may have been blind to low levels of organics.” To rebut this assertion, the Viking experiment, test data, and results on Mars are reviewed, and the fallacies in the design, execution, and interpretation of the new experiments presented by Navarro-Gonzalez et al. are critically examined.

The development of mass spectrometric methodologies for the sequencing of peptides and proteins are recounted. Early strategies for the determination of very large proteins based on a combination of nucleotide sequencing and mass spectrometric amino acid sequencing are described and their historical significance to the new field of proteomics is outlined.

The development of mass spectrometric methods in peptide and protein chemistry in the author's laboratory is reviewed, from the first determination of the amino acid sequence of small peptides in the late 1950s to its use for the determination of the primary structure of large proteins by a combination of mass spectrometry and DNA sequencing in the late 1980s. This article is part of a Special Issue entitled: 20years of Proteomics in memory of Viatliano Pallini. Guest Editors: Luca Bini, Juan J. Calvete, Natacha Turck, Denis Hochstrasser and Jean-Charles Sanchez.

Heparin has been used as a clinical anticoagulant for more than 50 years, making it one of the most effective pharmacological agents known. Much of heparin's activity can be traced to its ability to bind antithrombin III (AT-III). Low molecular weight heparin (LMWH), derived from heparin by its controlled breakdown, maintains much of the antithrombotic activity of heparin without many of the serious side effects. The clinical significance of LMWH has highlighted the need to understand and develop chemical or enzymatic means to generate it. The primary enzymatic tools used for the production of LMWH are the heparinases from Flavobacterium heparinum , specifically heparinases I and II. Using pentasaccharide and hexasaccharide model compounds, we show that heparinases I and II, but not heparinase III, cleave the AT-III binding site, leaving only a partially intact site. Furthermore, we show herein that glucosamine 3-O sulfation at the reducing end of a glycosidic linkage imparts resistance to heparinase I, II, and III cleavage. Finally, we examine the biological and pharmacological consequences of a heparin oligosaccharide that contains only a partial AT-III binding site. We show that such an oligosaccharide lacks some of the functional attributes of heparin- and heparan sulfate-like glycosaminoglycans containing an intact AT-III site.

The extent and distribution of N-glycosylation and the nature of most of the disulfide bond linkages were determined for bovine lactoperoxidase through proteolytic and glycolytic digestions combined with matrix-assisted laser desorption/ionization mass spectrometric analysis. In addition, 98% of the primary sequence of the protein was confirmed. All five of the asparagines present in sequons were found to be glycosylated, predominantly by high mannose and complex structures. Six disulfide bonds were assigned, including Cys 32-Cys 45, Cys 146-Cys 156, Cys 150-Cys 174, Cys 254-Cys 265, Cys 473-Cys 530 and Cys 571-Cys 596.

The extent of N-glycosylation of yeast external invertase at each of the 14 potential sites was determined by the combination of proteolytic digestions and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI/TOF-MS). The average molecular mass of the intact external invertase was determined as 97 kDa by MALDI/TOF-MS. The intact protein was digested with trypsin, Lys-C and Asp-N, followed by high-performance liquid chromatographic separation. The proteolytic digests were analyzed by MALDI/MS screening for the glycopeptides. The glycopeptides were then treated with peptide:N-glycosidase F (PNGase F) and/or endo-beta-N-acetylglucosaminidase (Endo H) and the molecular mass of the deglycosylated peptide was determined by MALDI/MS and matched with the peptide predicted by a computer program. The sequences of some peptides or deglycosylated peptides were identified by the MALDI post-source decay technique. The size of the oligosaccharide, the degree of glycosylation and the distribution of the oligosaccharides at each individual potential glycosylation site were characterized. This information goes for beyond previously published data and sometimes differs from them. During this study, the amino acid sequence originally derived from the DNA sequence of the gene coding for invertase was also verified and it was found that this protein when expressed from SUC2 gene might be created as more than one sequence which differ by a few amino acid substitutions (Asn58--Thr, Asn65--His and Val412--Ala).

Difficulties in determining composition and sequence of glycosaminoglycans, such as those related to heparin, have limited the investigation of these biologically important molecules. Here, we report methodology, based on matrix-assisted laser desorption ionization MS and capillary electrophoresis, to follow the time course of the enzymatic degradation of heparin-like glycosaminoglycans through the intermediate stages to the end products. MS allows the determination of the molecular weights of the sulfated carbohydrate intermediates and their approximate relative abundances at different time points of the experiment. Capillary electrophoresis subsequently is used to follow more accurately the abundance of the components and also to measure sulfated disaccharides for which MS is not well applicable. For those substrates that produce identical or isomeric intermediates, the reducing end of the carbohydrate chain was converted to the semicarbazone. This conversion increases the molecular weight of all products retaining the reducing terminus by the “mass tag” (in this case 56 Da) and thus distinguishes them from other products. A few picomoles of heparin-derived, sulfated hexa- to decasaccharides of known structure were subjected to heparinase I digestion and analyzed. The results indicate that the enzyme acts primarily exolytically and in a processive mode. The methodology described should be equally useful for other enzymes, including those modified by site-directed mutagenesis, and may lead to the development of an approach to the sequencing of complex glycosaminoglycans.

This article is © 1965 American Chemical Society, and permission has not been granted for it to be reproduced in electronic form.

Large, spherical dendrimers based on carbosilane chemistry that carry four, eight and 16 end-groups, respectively, varying from non-polar to basic and to fixed positive or negative charges were investigated by matrix-assisted laser desorption ionization mass spectrometry. It was found that those ending in tertiary amino groups are easily detected as their [M + H]+ ions, whereas non-polar dendrimers ending in chloroalkyl groups required the addition of silver trifluoroacetate to produce [M + Ag]+ ions. Those with SO3Na end-groups could be analyzed as [M − H]− ions after conversion to the corresponding poly-ammonium salts. However, dendrimers terminating in four or eight trimethylammonium chlorides and iodides, respectively, generated complexes of the tetra- and octa-cations with three and seven, respectively, matrix anions, such as 2,5-dihydroxybenzoate or 5-chlorosalicylate anions. From the m / z value of the resulting mono-positively charged ion, the molecular weight of the dendrimer polycation can be deduced.

The sites of glycosylation of Chinese hamster ovary cell expressed testicular angiotensin-converting enzyme (tACE) have been determined by matrix-assisted laser desorption ionization/time of flight/mass spectrometry of peptides generated by proteolytic and cyanogen bromide digestion. Two of the seven potential N-linked glycosylation sites, Asn90 and Asn109, were found to be fully glycosylated by analysis of peptides before and after treatment with a series of glycosidases and with endoproteinase Asp-N. The mass spectra of the glycopeptides exhibit characteristic clusters of peaks which indicate the N-linked glycans in tACE to be mostly of the biantennary, fucosylated complex type. This structural information was used to demonstrate that three other sites, Asn155, Asn337, and Asn586, are partially glycosylated, whereas Asn72 appears to be fully glycosylated. The only potential site that was not modified is Asn620. Sequence analysis of tryptic peptides obtained from somatic ACE (human kidney) identified six glycosylated and one unglycosylated Asn. Only one of these glycosylation sites had a counterpart in tACE. Comparison of the two proteins reveals a pattern in which amino-terminal N-linked sites are preferred. The functional significance of glycosylation was examined with a tACE mutant lacking the O-glycan-rich first amino-terminal 36 residues and truncated at Ser625. When expressed in the presence of the alpha-glucosidase I inhibitor N-butyldeoxynojirimycin and treated with endoglycosidase H to remove all but the terminal N-acetylglucosamine residues, it retained full enzymatic activity, was electrophoretically homogeneous, and is a good candidate for crystallographic studies.

The high-energy collision-induced dissociation (CID) spectra of a number of peptides containing S-β-(4-pyridylethyl)-, S-carbamidomethyl- and S-carboxymethylcysteine have been generated and several hitherto undescribed fragmentation pathways identified. For peptides containing pyridylethylcysteine at the n th position, these include x n − 105, a n +1 + 1 − 106, a n +1 + 1 − 138, z n +1 + 1 − 106, z n +1 + 1 − 138, and combined side chain loss ions. The analogous fragment ions were also observed in the spectra of peptides containing other S-alkylcysteines. Also, two fragmentations unique to pyridylethylcysteine-containing peptides, resulting in the formation of b n + ≥ 1 - R C + H and CO + peC ions, were studied. Precursor scanning, exact mass measurement and CID of the appropriate source-generated fragment ions have been employed to identify the origin of these ions and to facilitate the proposal of mechanisms for their formation.

Incubation of the C225S mutant of the R1 subunit of ribonucleotide reductase from Escherichia coli with the R2 subunit and nucleoside diphosphates leads to fragmentation of the polypeptide backbone of R1 [Mao, S. S., Holler, T. P., Bollinger, J.M., Jr., Yu, G. X., Johnston, M.I., & Stubbe, J. (1992) Biochemistry 31, 9744--9751]. The 26 and 60 kDa cleavage fragments were purified to homogeneity. The 26 kDa polypeptide was digested with Lys-C, and the peptides were partially purified by RP-HPLC. Mass spectrometric analysis (MALDI-TOF) of the HPLC fractions allowed the identification of the C-terminal peptide. The molecular mass of this peptide (2176) revealed that serine-224 constitutes its C-terminus, and further analysis of the distribution of its monoisotopic masses by FAB-MS indicated that Ser224 possesses a carboxamide rather than a carboxylate group. Treatment of the 60 kDa cleavage fragment with cyanogen bromide and subsequent MALDI-TOF analysis of the partially RP-HPLC purified peptides yielded a fraction containing its N-terminal peptide. This peptide was digested with trypsin, and the digestion mixture was purified by HPLC. Analysis of the fractions by MALDI-TOF identified the N-terminal peptide and determined a mass of 2222. This mass suggested valine 226 was the N-terminal residue (modified by an adduct of 28 mass units). Larger amounts of the C-terminal tetrapeptide of the 60 kDa fragment (V226LIE229) were obtained by complete digestion of the crude reaction mixture with endoproteinase Glu-C. The peptide mixture was then purified on an immunoadsorbent column containing immobilized antibodies raised against a synthetic peptide with the sequence KVLIE. After elution of the affinity-bound peptide, it was analyzed by CID-MS verifying that an adduct of 28 mass units was attached to valine 226. These results indicated that the amino group of Val226 is formylated. The localization of the residues at the cleavage site of C225SR1 provides a biochemical identification of the active site region of the R1 subunit of RDPR from E.coli. The details of the mechanism of cleavage remain to be elucidated.

The extent of charging (or protonation) during the electrospray ionization process has been examined for a series of specifically constructed arginine-rich peptides, which differ in structure by the length of the peptide chain and the number and proximity of arginine residues. It has been found that although a small peptide of the series will protonate fully, supporting a charge on each arginine side-chain, the same charging behavior is not observed for larger peptides with the same repeating primary structure. Furthermore, no significant increase in the extent of charging was observed as the length of the peptide chain, or the distance between potential charge-bearing sites, was increased. The apparent sites of protonation in the [M + nH]n+ peptide ions have been examined for several representative peptides based on the extent of protonation compared to that of structurally related peptides, and their dissociation behavior. Despite the potential for proton migration during the collisional activation event, the fragmentation pattern of the peptide ions studied suggests that the charge-bearing protons are reasonably localized at the time of dissociation commensurate with our previous observations for singly and multiply charged peptide ions. The charging behavior of the model peptides is discussed in the context of a reported mechanism for the electrospray ionization process.

Fixed-charge derivatives have been used to direct the fragmentation pattern of high energy collision-induced dissociation tandem mass spectra for several years. It has been noted that a fixed-charge placed at a terminus of a peptide will simplify the pattern of fragment ions that are produced in collision-induced dissociation. Trimethylammoniumacetyl, dimethyloctylammoniumacetyl, and triphenylphosphoniumethyl derivatives have been cited in the literature for this purpose and many other structures are possible. This work compares the cited derivatives as well as some new structures. The criteria used include the ease of synthesis and purification of the derivatized peptide and the effects of the derivative on the peptide sequence fragment ion yield and ionization efficiency. The trimethylammoniumacetyl derivative is concluded to be the most practical for general use, whereas the dimethyloctylammoniumacetyl derivative is found to be desirable for use with hydrophilic peptides.

Two methionine-specific sequence ions are described that are formed upon the dissociation of protonated methionine-containing peptides at high (keV) collision energies. These ions, hiherto unassigned, are denoted an + 1 − CH3S⋅ and zn + 1 − CH3S⋅. Precursor ion scans reveal that these species originate from the an + 1 and zn + 1 ions, respectively, where methionine resides at position n-1 relative to the N- or C-terminus. A comparison of the collision-induced dissociation (CID) spectra for several related peptides demonstrates that methionine residues are involved in the genesis of these species and mechanisms for their formation are proposed. Several structural features of a peptide that are important to the formation of these species are described. The importance of identifying these ions in order to correctly sequence a methionine-containing peptide based on its CID spectrum is illustrated.

The events that led to the establishment of organic and biochemical mass spectrometry at MIT by the author in 1958, and its growth over the past three and one-half decades are briefly chronicled. A major emphasis is placed on the work with graduate students and postdoctoral researchers who were educated in the field and in turn further contributed to the training of others. An attempt is made at the construction of a genealogy encompassing −7 to +2 generations (the author representing 0).

Thioltransferase (glutaredoxin) was purified from human red blood cells essentially as described previously (Mieyal JJ et al., 1991a, Biochemistry 30:6088-6097). The primary sequence of the HPLC-pure enzyme was determined by tandem mass spectrometry and found to represent a 105-amino acid protein of molecular weight 11,688 Da. The physicochemical and catalytic properties of this enzyme are common to the group of proteins called glutaredoxins among the family of thiol:disulfide oxidoreductases that also includes thioredoxin and protein disulfide isomerase. Although this human red blood cell glutaredoxin (hRBC Grx) is highly homologous to the 3 other mammalian Grx proteins whose sequences are known (calf thymus, rabbit bone marrow, and pig liver), there are a number of significant differences. Most notably an additional cysteine residue (Cys-7) occurs near the N-terminus of the human enzyme in place of a serine residue in the other proteins. In addition, residue 51 of hRBC Grx displayed a mixture of Asp and Asn. This result is consistent with isoelectric focusing analysis, which revealed 2 distinct bands for either the oxidized or reduced forms of the protein. Because the enzyme was prepared from blood combined from a number of individual donors, it is not clear whether this Asp/Asn ambiguity represents inter-individual variation, gene duplication, or a deamidation artifact of purification.

A novel matrix substance, 2-(4-hydroxyphenylazo)benzoic acid, or HABA, has been found to be very advantageous for matrix-assisted ultraviolet laser desorption ionization mass spectrometry. This compound has been successfully used for the desorption of peptides, proteins, and glycoproteins up to approximately 250 kDa. For these materials, the most abundant analyte-related peaks correspond to [M + H]+ ions and multiply protonated molecules. Comparisons with sinapic acid, 2,5-dihydroxybenzoic acid, and α-cyano-4-hydroxycinnamic acid indicate that the new matrix provides comparable sensitivity for peptides and smaller proteins but results in better sensitivity for larger proteins and glycoproteins in protein mixtures. Other matrices discriminate against the higher mass components in these cases. Somewhat reduced mass resolution has been found for smaller proteins, but for larger proteins and glycoproteins the best mass resolution can often be obtained with the new matrix. For other classes of compounds that form ions predominantly via cation attachment, at least as good sensitivity and even better resolution have been obtained. Derivatized glycolipids and synthetic polymers have been studied in detail. For the analysis of many synthetic polymers, the best performance in terms of sensitivity and mass resolution has been observed with HABA matrix. Mass resolution was higher for cation adducts than for the protonated peptide molecules in the same mass range. The new matrix exhibits grealty extended (in time) analyte ion production and reproducibility. Owing to the uniform sample surface with this matrix, barely any spatial variation of the ion signal could be observed. In addition, many hundreds of single-shot mass spectra could be accumulated from the same spot, even for larger proteins.

Acetylation of the animo groups (N-terminus and lysine) of proteins before enzymatic or chemical cleavage was explored as an approach to provide additional information in the course of the determination of amino acid sequences. The major advantage is the ability to differentiate glutamine from lysine, because only the latter is acetylated and thus increases in mass by 42 Da. Horse heart cytochrome c could be fully N-actetylated and even on prolonged digestion with chymotrypsin underwent very little tryptic cleavage, in contrast to the native protein where this side reaction is extensive. Sperm whale myoglobin is more difficult to acetylate, but even at 40%–50% average acetylation, all 19 lysines could be identified unambiguously. A proteolytic digest of acetylated protein is thus a useful component of strategies for the determination of the primary structure of proteins by tandem mass spectrometry.

It has been observed that in the analysis of peptide mixtures by fast atom bombardment mass spectrometry some of the components are not detected and, usually, the ion yields of protonated peptide molecules do not reflect their relative molar abundance. This phenomenon has been attributed to the suppression of ionization of some of the mixture components by other components. It has been found that a change in the sample probe surface temperature, effected with a thermoelectrically-cooled probe, eliminates the suppression effects and, in some instances, even reverses them.

Four variants of human hemoglobin have been structurally characterized by mass spectrometry. The molecular weight changes caused by the mutations were −58, + 14, + 1, and −1 u respectively. The detection of the last two mutations was based on the observation of a distorted isotope pattern for the (M + H) + ion of the proteolytic peptides encompassing the site of mutation, because the mutants are always accompanied by the normal protein chain. For those cases where the molecular weight measurements alone do not unambiguously determine the nature and location of the modification, tandem mass spectrometry had to be used. Of the four variants, three proved to have been reported previously: α68Asn→Asp (Hb Ube-II), β22Glu→Ala (Hb G Coushatta) and β43Glu→Gln (Hb Hoshida). The fourth variant was found to be due to a β79Asp→Glu mutation that had not been discovered before and was named Hb Miyazono.

A conventional fast-atom bombardment (FAB) ion source was used to achieve matrix-assisted laser desorption (MALD) in a high-mass, double-focusing, magnetic mass spectrometer. The pulsed ion signals generated by irradiation of a mixture of sample and matrix (2,5-dihydroxybenzoic acid) with either a XeF excimer laser (353 nm) or a nitrogen laser (337 nm) were recorded with a focal-plane detector. A resolution (full-width at half maximum) of 4500 was achieved at m/z 1347.7 (the peptide substance P), 2500 for CsI cluster ions at m/z 10,005.7, and 1250 for the isotope cluster of the small protein cytochrome c (horse) [M+H]+ = m/z 12,360 (average). Sensitivity is demonstrated with 11 fmol of substance P. A survey scan is taken to locate the m/z of the sample molecular ion. The segment that contains the sample can then be integrated for a longer time to produce a better signal-to-noise ratio. In addition to higher sensitivity and lower matrix interference, the advantage of MALD over FAB is the former's lower susceptibility to the presence of salts, and competition between hydrophobic and hydrophilic components of a mixture. This feature is demonstrated by the complete MALD spectrum of a crude partial tryptic digest of sperm-whale apomyoglobin, containing 24 peptides, representing the entire sequence of this protein.

The amino acid sequence of a protease inhibitor isolated from the hemolymph of Sarcophaga bullata larvae was determined by tandem mass spectrometry. Homology considerations with respect to other protease inhibitors with known primary structures assisted in the choice of the procedure followed in the sequence determination and in the alignment of the various peptides obtained from specific chemical cleavage at cysteines and enzyme digests of the S. bullata protease inhibitor. The resulting sequence of 57 residues is as follows: Val Asp Lys Ser Ala Cys Leu Gln Pro Lys Glu Val Gly Pro Cys Arg Lys Ser Asp Phe Val Phe Phe Tyr Asn Ala Asp Thr Lys Ala Cys Glu Glu Phe Leu Tyr Gly Gly Cys Arg Gly Asn Asp Asn Arg Phe Asn Thr Lys Glu Glu Cys Glu Lys Leu Cys Leu.

Myoglobins from horse heart muscle, horse skeletal muscle and sperm whale are widely used as calibration standards or test compounds for various mass spectrometric methodologies. In all such cases reported in the literature, a molecular weight value is used (16,950.5 and 17,199, respectively) which is based on the assumption that amino acid 122 in this 153 amino-acid-long protein is asparagine, overlooking a published suggestion that it is aspartic acid instead. Since the mass assignment accuracy for matrix-assisted laser desorption mass spectrometry is reported to be +/- 0.01% and for electrospray ionization +/- 0.0025%, and error of one mass unit in approximately 17,000 would be significant. The mass-to-charge ratio of ions of the tryptic peptide encompassing amino acid 122 derived from commercially available horse heart and horse skeletal myoglobins, the apomyoglobin of the latter, and the tryptic and chymotryptic peptide of sperm whale myoglobin proved that in both proteins amino acid 122 is indeed aspartic acid, rather than asparagine. This finding was further confirmed by the collision-induced dissociation spectra of the [M + H]+ ions of the tryptic peptides from the horse myoglobins and the chymotriptic peptide from sperm whale myoglobin. Thus, the correct molecular weight of horse myoglobin is 16,951.49 and that of the sperm whale protein is 17,199.91.

Incorporation of a pair of quadrupole lenses and one octapole lens after the second magnetic field of a JEOL HX110/HX110 tandem mass spectrometer permits varying the dispersion and therefore the mass range that can be focused on a 2 in array detector. This range can be adjusted from a value of 1: 1.04 to 1: 1.3 for the ratio of low m/z to high m/z falling at the extremes of the detector. The ratio is governed by the d.c. potentials on the lenses which also changes the angle of the resulting focal plane and thus requires that the position of the array detector must be changed accordingly. In the system described, all these variables are under computer control and can be adjusted within 1 s, thus allowing for a change even during the recording of a spectrum if one wishes to minimize the number of segments to be recorded to cover the entire spectrum while still maintaining unit mass resolution. The variable range array detector is useful for the recording of normal mass spectra as well as those generated upon collision induced decomposition (CID) of precursor ions. In the latter case, the refocusing of the ion beams through the quadrupole lens system is more complex but the software controls the lens potentials accordingly. The system has been used to record the normal or CID spectra of peptides, either from a static sample or eluting from a high performance liquid chromatograph.

For the differentiation of leucine and isoleucine in a peptide by high energy collision-induced dissociation (CID), it is generally required that there is a basic amino acid present at or near either the C-terminus or the N-terminus of the peptide. In these cases, fragmentation or the β,γ bond of the side chain occurs, generating ions designated w n or d n that permit the differentiation of these isomeric amino acids. While trypsin and Endo Lys-C generate peptides with a basic C-terminal amino acid, other enzymes cleave at neutral or acidic amino acids and may thus produce proteolytic peptides that do not contain any basic amino acids. For these, a microderivatization method has been developed that places a fixed positive charge at the C-terminus. It involves exposure of the peptide(s) deposited on the inner wall of a capillary tube, first to chloroacetyl chloride vapor and then to trimethylamine and water vapors. This two-step reaction attaches a trimethylammonium acetyl (TMA) moiety to the N-terminal amino group of the peptide. The CID spectra of these derivatives are very simple, exhibit the same characteristics (including abundant d n ions) as peptides bearing the strongly basic arginine at the N-terminus, and thus permit the differentiation of leucine from isoleucine. The reaction can be carried out at the sub-nanomole level.

Heparin- and heparan sulfate-like glycosaminoglycans (HLGAGs) represent an important class of molecules that interact with and modulate the activity of growth factors, enzymes, and morphogens. Of the many biological functions for this class of molecules, one of its most important functions is its interaction with antithrombin III (AT-III). AT-III binding to a specific heparin pentasaccharide sequence, containing an unusual 3-O sulfate on a N-sulfated, 6-O sulfated glucosamine, increases 1,000-fold AT-III's ability to inhibit specific proteases in the coagulation cascade. In this manner, HLGAGs play an important biological and pharmacological role in the modulation of blood clotting. Recently, a sequencing methodology was developed to further structure-function relationships of this important class of molecules. This methodology combines a property-encoded nomenclature scheme to handle the large information content (properties) of HLGAGs, with matrix-assisted laser desorption ionization MS and enzymatic and chemical degradation as experimental constraints to rapidly sequence picomole quantities of HLGAG oligosaccharides. Using the above property-encoded nomenclature-matrix-assisted laser desorption ionization approach, we found that the sequence of the decasaccharide used in this study is ΔU 2S H NS,6S I 2S H NS,6S I 2S H NS,6S IH NAc,6S GH NS,3S,6S (±DDD4–7). We confirmed our results by using integral glycan sequencing and one-dimensional proton NMR. Furthermore, we show that this approach is flexible and is able to derive sequence information on an oligosaccharide mixture. Thus, this methodology will make possible both the analysis of other unusual sequences in HLGAGs with important biological activity as well as provide the basis for the structural analysis of these pharamacologically important group of heparin/heparan sulfates.

Background: The angiogenesis inhibitors TNP-470 and ovalicin potently suppress endothelial cell growth. Both drugs also specifically inhibit methionine aminopeptidase 2 (MetAP2) in vitro . Inhibition of MetAP2 and changes in initiator methionine removal in drug-treated endothelial cells have not been demonstrated, however. Results: Concentrations of TNP-470 sufficient to inactivate MetAP2 in intact endothelial cells were comparable to those that inhibited cell proliferation, suggesting that MetAP2 inhibition by TNP-470 underlies the ability of the drug to inhibit cell growth. Both drug-sensitive and drug-insensitive cell lines express MetAP1 and MetAP2, indicating that drug sensitivity in mammalian cells is not simply due to the absence of compensating MetAP activity. With a single exception, detectable protein N-myristoylation is unaffected in sensitive endothelial cells treated with TNP-470, so MetAP1 activity can generally compensate when MetAP2 is inactive. Analysis of total protein extracts from cells pulse-labeled with [ 35 S]-methionine following TNP-470 treatment revealed changes in the migration of several newly synthesized proteins. Two of these proteins were identified as glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and cyclophilin A. Purification and amino-terminal sequencing of GAPDH from TNP-470-treated cells revealed partial retention of its initiator methionine, indicating that methionine removal from some, but not all, proteins is affected by MetAP2 inactivation. Conclusions: Amino-terminal processing defects occur in cells treated with TNP-470, indicating that inhibition of MetAP2 by the drug occurs in intact cells. This work renders plausible a mechanism for growth inhibition by TNP-470 as a consequence of initiator methionine retention, leading to the inactivation of as yet unidentified proteins essential for endothelial cell growth.

The feasibility of using photodissociation of protonated peptide molecules to sequence specific fragment ions with a 193-nm pulsed laser beam in a magnetic deflection tandem mass spectrometer of EBEB configuration was demonstrated. Although the short pulse (15 ns) and low repetition rate (100 Hz) of the excimer laser permitted the irradiation of only ∼ 0.02% of the (M + H)+ ions exiting MS-1, a photon-induced decomposition spectrum of the heptapeptide angiotensio III (M r 930.5) was produced that was practically the same (but with better signal-to-noise ratio) as that generated by collision-activated dissociation at the same low duty cycle. Because of the low and pulsed fragment ion currents, an array detector was used to record the spectra. A dependence between laser power and abundance of fragment ions was observed (increased power increases the relative abundance of ions of low mass). Laser power was varied from 6 to 80 mJ. Formation of fragment ions from a large peptide (melittin, M, 2844.75) was also observed. The results permit the design of modifications that may increase the fragment ion yield to 10% or higher, which would make photon-induced decomposition a useful method for magnetic deflection mass spectrometers.

Heparin-like glycosaminoglycans, acidic complex polysaccharides present on cell surfaces and in the extracellular matrix, regulate important physiological processes such as anticoagulation and angiogenesis. Heparin-like glycosaminoglycan degrading enzymes or heparinases are powerful tools that have enabled the elucidation of important biological properties of heparin-like glycosaminoglycans in vitro and in vivo . With an overall goal of developing an approach to sequence heparin-like glycosaminoglycans using the heparinases, we recently have elaborated a mass spectrometry methodology to elucidate the mechanism of depolymerization of heparin-like glycosaminoglycans by heparinase I. In this study, we investigate the mechanism of depolymerization of heparin-like glycosaminoglycans by heparinase II, which possesses the broadest known substrate specificity of the heparinases. We show here that heparinase II cleaves heparin-like glycosaminoglycans endolytically in a nonrandom manner. In addition, we show that heparinase II has two distinct active sites and provide evidence that one of the active sites is heparinase I-like, cleaving at hexosamine–sulfated iduronate linkages, whereas the other is presumably heparinase III-like, cleaving at hexosamine–glucuronate linkages. Elucidation of the mechanism of depolymerization of heparin-like glycosaminoglycans by the heparinases and mutant heparinases could pave the way to the development of much needed methods to sequence heparin-like glycosaminoglycans.

Heparinase I from Flavobacterium heparinum has important uses for elucidating the complex sequence heterogeneity of heparin-like glycosaminoglycans (HLGAGs). Understanding the biological function of HLGAGs has been impaired by the limited methods for analysis of pure or mixed oligosaccharide fragments. Here, we use methodologies involving MS and capillary electrophoresis to investigate the sequence of events during heparinase I depolymerization of HLGAGs. In an initial step, heparinase I preferentially cleaves exolytically at the nonreducing terminal linkage of the HLGAG chain, although it also cleaves internal linkages at a detectable rate. In a second step, heparinase I has a strong preference for cleaving the same substrate molecule processively, i.e., to cleave the next site toward the reducing end of the HLGAG chain. Computer simulation showed that the experimental results presented here from analysis of oligosaccharide degradation were consistent with literature data for degradation of polymeric HLGAG by heparinase I. This study presents direct evidence for a predominantly exolytic and processive mechanism of depolymerization of HLGAG by heparinase I.

Glycosaminoglycans (GAGs) are a class of highly acidic polysaccharides that are found in the extracellular matrix and in proteoglycans, which are cell membrane-bound proteins that are post-translationally modified with GAGs (1, 2). Analysis of these compounds by classical methods has been difficult due to the heterogeneous and polyanionic nature of GAGs. However, by mixing GAG oligosaccharides with basic peptides to form a noncovalent complex, it is possible to accurately determine their molecular weight using matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) (3, 4). Such molecular weight measurements provide information on the length of the carbohydrate chain and the degree of sulfation. In addition, mass spectrometric measurements on enzymatic and chemical degradation products of isolated compounds may yield the entire structure of the analyte of interest. Figure 5.1 represents the structure of a heparin fragment.

Human testicular angiotensin-converting enzyme (tACE) is an extracellular protein that contains seven cysteine residues. The cysteines occur in a sequential distribution that is precisely mimicked in the tACE from rabbit and mouse, and in both domains of all known species of somatic ACE. One of the cysteines in human tACE, Cys496, is present in the reduced form as shown by labeling it with 5-[[2-(iodoacetyl)amino]ethylamino]naphthalene-1-sulfonic acid, isolating the fluorescent peptide from enzymatic digests by HPLC, and analyzing its sequence by matrix-assisted laser desorption ionization-mass spectrometry (MALDI-MS). Thiol reagents have no significant effect on the activity of tACE, indicating that this Cys is not involved in catalysis. The other six cysteines exist as three disulfides. Mass spectral analysis of cyanogen bromide peptides has established that the cystine connectivities follow a nearest-neighbor, aabbcc, pattern i.e., Cys152-Cys158, Cys352-Cys370, and Cys538-Cys550, in which the disulfides form three small loops of five, 17, and 11 residues, respectively. Although these disulfide loops constitute less than 5% of the total sequence of the protein, they contribute to the overall structural stabilization of tACE.

The characterization of the source of the odor in the human axillary region is not only of commercial interest but is also important biologically because axillary extracts can alter the length and timing of the female menstrual cycle. In males, the most abundant odor component is known to be E-3-methyl-2-hexenoic acid (E-3M2H), which is liberated from nonodorous apocrine secretions by axillary microorganisms. Recently, it was found that in the apocrine gland secretions, 3M2H is carried to the skin surface bound to two proteins, apocrine secretion odor-binding proteins 1 and 2 (ASOB1 and ASOB2) with apparent molecular masses of 45 kDa and 26 kDa, respectively. To better understand the formation of axillary odors and the structural relationship between 3M2H and its carrier protein, the amino acid sequence and glycosylation pattern of ASOB2 were determined by mass spectrometry. The ASOB2 protein was identified as apolipoprotein D (apoD), a known member of the alpha2mu-microglobulin superfamily of carrier proteins also known as lipocalins. The pattern of glycosylation for axillary apoD differs from that reported for plasma apoD, suggesting different sites of expression for the two glycoproteins. In situ hybridization of an oligonucleotide probe against apoD mRNA with axillary tissue demonstrates that the message for synthesis of this protein is specific to the apocrine glands. These results suggest a remarkable similarity between human axillary secretions and nonhuman mammalian odor sources, where lipocalins have been shown to carry the odoriferous signals used in pheromonal communication.

The amino acid sequences of a number of closely related proteins ("napin") isolated from Brassica napus were determined by mass spectrometry without prior separation into individual components. Some of these proteins correspond to those previously deduced (napA, BngNAP1, and gNa), chiefly from DNA sequences. Others were found to differ to a varying extent (BngNAP1', BngNAP1A, BngNAP1B, BngNAP1C, gNa', and gNaA). The short chains of gNa and gNa' and of BngNAP1 and BngNAP1' differ by the replacement of N-terminal proline by pyroglutamic acid; the long chains of gNaA and BngNAP1B contain a six amino acid stretch, MQGQQM, which is present in gNa (according to its DNA sequence) but absent from BngNAP1 and BngNAP1C. These alternations of sequences between napin isoforms are most likely due to homologous recombination of the genetic material, but some of the changes may also be due to RNA editing. The amino acids that follow the untruncated C termini of those napin chains for which the DNA sequences are known (napA, BngNAP1, and gNa) are aromatic amino acids. This suggests that the processing of the proprotein leading to the C termini of the two chains is due to the action of a protease that specifically cleaves a G/S-F/Y/W bond.

Inosine 5'-monophosphate dehydrogenase (IMPDH) is the rate-limiting enzyme in de novo guanine nucleotide biosynthesis. IMPDH converts inosine 5'-monophosphate (IMP) to xanthosine 5'-monophosphate (XMP) with concomitant conversion of NAD+ to NADH. The antiviral agent 5-ethynyl-1-beta-D-ribofuranosylimidazole-4-carboxamide (EICAR) is believed to inhibit IMPDH by forming an active metabolite, the 5'-monophosphate EICARMP. The experiments reported here demonstrate that EICARMP irreversibly inactivates both human type II and Escherichia coli IMPDH. IMPDH is protected from EICARMP inactivation by IMP, but not by NAD+. Further, denaturation/renaturation of the EICARMP-inactivated enzyme did not restore enzyme activity, which indicates that EICARMP forms a covalent adduct with IMPDH. EICARMP was successfully used to titrate the active sites of IMPDH; these experiments demonstrate that four active sites are present in an IMPDH tetramer. Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry of native E. coli IMPDH established that protein translation initiates at the third ATG of the DNA sequence. Thus, the E. coli IMPDH monomer is only 488 amino acids long and contains five instead of six cysteines. In addition, MALDI-TOF mass spectrometry showed that EICARMP is covalently bound to Cys-305 (Cys-331 in human type II IMPDH numbering), suggesting that Cys-305 functions as a nucleophile in the IMPDH reaction. The inactivation of the E. coli enzyme is a single-step reaction with kon = 1.94 x 10(4) M-1 s-1. In contrast, the inactivation of human type II IMPDH involves a two-step mechanism where Ki = 16 microM, k2 = 2.7 x 10(-2) s-1 and kon = 1.7 x 10(3) M-1 s-1. These results demonstrate that significant differences exist between bacterial and human IMPDH and suggest that this enzyme may be a target for antibiotic drugs.

Molecular weights of heparin-derived oligosaccharides ranging from disaccharides to hexadecasaccharides have been determined by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. While these compounds ionize poorly or not at all when used as such, a strong signal can be obtained of their ionic complexes formed with a basic peptide or protein. The molecular weight of the sulfated oligosaccharide is determined by subtracting the mass of the basic component from that of the complex. Optimization of the experimental conditions resulted in sub-picomole sensitivity, in the elimination of sulfate loss and of the interference from attachment of inorganic cations. Synthetic peptides (Arg-Gly)10 and (Arg-Gly)15 were specifically designed as complexing agents for synthetic and natural heparin fragments up to decasaccharides. Accurate molecular weight determination on chemically homogeneous oligosaccharides (+/- 0.05%) unambiguously identified the number of saccharide units, and the number of O,N-sulfate and N-acetyl groups. For oligosaccharides larger than decasaccharides, a small basic protein, angiogenin (M(r) = 14,120), was used to form the complex (an inhomogeneous hexadecasaccharide fraction was the largest available for this study). For inhomogeneous samples larger than decasaccharides, the mass accuracy is lower (+/- 0.2-0.3%) but still suffices to determine the number of saccharide units present and to estimate the number of sulfate groups, except it is no longer possible to differentiate one sulfate from two N-acetyl groups (delta = 4 Da). However, taking into account known regularities of sulfation and acetylation, the specificity of heparin lyases and chemical degradation steps, the method promises to contribute significantly to the determination of the primary structure of heparin and other sulfated glycosaminoglycans.

A variety of chemically modified oligonucleotides have been studied by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) in the negative ion mode. These include oligonucleotides containing modified bases, such as uracil glycol, bromoguanine, O6-butylguanine, as well as oligonucleotides in which the phosphodiester groups had been replaced by other functional groups, such as phosphorothioates. With the linear TOF mass spectrometer, there is no or very little fragmentation observed, and the determination of the molecular weight by MALDI-TOFMS offers a convenient way for identifying/confirming the presence of the modification. With internal calibration, a mass accuracy of 0.01% can be achieved. Such mass accuracy makes it possible to directly differentiate a small uridine-containing oligonucleotide from its cytidine-containing analogue. Because of factors such as sample inhomogeneity, laser output fluctuation, and the dynamic range of the detector, quantitation by MALDI-TOFMS has been difficult. Nevertheless, semiquantitative information can be obtained for those analytes that are closely related in structure. Monitoring the products of the synthesis of monophosphorothioated oligoribonucleotide 16-mers by MALDI-TOFMS revealed that the sulfur atom in the phosphorothioate group can be replaced by an oxygen atom during the succeeding introduction of phosphodiester groups. The earlier the phosphorothioate group is introduced during the synthesis of the 16-mer, the greater is the extent of sulfur to oxygen replacement.

Highly acidic compounds that are difficult to ionize by matrix-assisted laser desorption ionization give excellent spectra when mixed with a basic peptide or protein to form a noncovalent complex. This phenomenon makes it possible to determine the molecular weights of polysulfated, -sulfonated, and -phosphorylated biomolecules such as cysteic acid-containing peptides, oligonucleotides, heparin-derived oligosaccharides, and suramin (a drug containing two trisulfonated naphthalene moieties). Peptides and small proteins rich in arginine were used as the basic components. The extent of complex formation correlates with the number of phosphate and sulfate groups in the acidic component and with the number of arginines in the basic component. Neither the acidic amino acid residue aspartic and glutamic acid nor the basic lysine and histidine contribute to complex formation. For oligonucleotides, histone H4 was found to be the best complexing agent investigated. The analytical utility of the complex formation is demonstrated by the molecular-mass determination of acidic compounds from 500 to 6000 Da at the picomole or sub-picomole level with an accuracy of +/- 0.1% or better and by the absence of alkali cation adducts.

The effect that charge state has on the collision-induced dissociation (CID) of peptide ions is examined in detail for several representative peptides under high-energy collision conditions. The CID spectra of singly and doubly charged precursor ions (generated by fast-atom bombardment and electrospray ionization, respectively) are compared for several peptides with similar primary structure. It is shown that for peptides that contain highly basic amino acids, the dissociation of doubly charged ions is strongly influenced by the position of these residues within the peptide and the general observations reported concerning the dissociation of singly charged ions can be extended to precursors with higher charge states. Based on the dissociation behavior of the doubly charged ions of these peptides, it is demonstrated that two charges can reside in close proximity in the precursor ions, overcoming possible repulsion effects, when favored by a high concentration of basic sites. In addition, this work illustrates that in the case of doubly charged ions, the charge state of some fragment ions can be determined directly from the mass-to-charge ratio assignments of the CID spectrum.

Polycyclic aromatic hydrocarbon (PAH) diolepoxides are known to covalently modify serum albumin and hemoglobin. Mass spectrometric techniques have proven quite useful in the characterization of the site of adduction on these proteins. To facilitate the study of PAH diolepoxide adducted peptides, model peptide adducts of benzo[a]pyrene-rans-7,8-dihydrodiol-9,10-epoxide [anti-BaP(9,10)DE] and benzo[a]anthracene-trans-8,9-dihydrodiol-10,11-epoxide [anti-BaA(10,11)DE] have been synthesized for the purpose of studying their high energy collision-induced dissociation tandem mass spectra. These spectra are dominated by ions produced from cleavage of the peptide—adduct bond with charge retention by the adducting moiety. Such ions allow for the facile identification of adducted peptides in a mixture by use of neutral loss scans. The peptide sequence can still be deduced from the data in most cases, and the site of adduction can be determined. For those peptide—adducts in which this is not possible, a charged derivative placed at the N-terminus simplifies the peptide fragmentation pattern and makes the spectrum more interpretable.